Other formats:
BibTeX
LaTeX
RIS
@inproceedings{1392124,
author = {Míč, Vladimír and Novák, David and Zezula, Pavel},
address = {Cham},
booktitle = {Similarity Search and Applications: 10th International Conference, SISAP 2017, Munich, Germany, October 4-6, 2017, Proceedings},
doi = {http://dx.doi.org/10.1007/978-3-319-68474-1_4},
editor = {Christian Beecks, Felix Borutta, Peer Kroger, Thomas Seidl},
keywords = {Similarity search; Metric space; Space transformation; Hamming space; sketch},
howpublished = {tištěná verze "print"},
language = {eng},
location = {Cham},
isbn = {978-3-319-68474-1},
pages = {53-63},
publisher = {Springer International Publishing},
title = {Sketches with Unbalanced Bits for Similarity Search},
year = {2017}
}
TY - JOUR
ID - 1392124
AU - Míč, Vladimír - Novák, David - Zezula, Pavel
PY - 2017
TI - Sketches with Unbalanced Bits for Similarity Search
PB - Springer International Publishing
CY - Cham
SN - 9783319684741
KW - Similarity search
KW - Metric space
KW - Space transformation
KW - Hamming space
KW - sketch
N2 - In order to accelerate efficiency of similarity search, compact bit-strings compared by the Hamming distance, so called sketches, have been proposed as a form of dimensionality reduction. To maximize the data compression and, at the same time, minimize the loss of information, sketches typically have the following two properties: (1) each bit divides datasets approximately in halves, i.e. bits are balanced, and (2) individual bits have low pairwise correlations, preferably zero. It has been shown that sketches with such properties are minimal with respect to the retained information. However, they are very difficult to index due to the dimensionality curse -- the range of distances is rather narrow and the distance to the nearest neighbour is high. We suggest to use sketches with unbalanced bits and we analyse their properties both analytically and experimentally. We show that such sketches can achieve practically the same quality of similarity search and they are much easier to index thanks to the decrease of distances to the nearest neighbours.
ER -
MÍČ, Vladimír, David NOVÁK and Pavel ZEZULA. Sketches with Unbalanced Bits for Similarity Search. In Christian Beecks, Felix Borutta, Peer Kroger, Thomas Seidl. \textit{Similarity Search and Applications: 10th International Conference, SISAP 2017, Munich, Germany, October 4-6, 2017, Proceedings}. Cham: Springer International Publishing, 2017, p.~53-63. ISBN~978-3-319-68474-1. Available from: https://dx.doi.org/10.1007/978-3-319-68474-1\_{}4.
|
